The candidate, Allegra Broft, M.D., requests support for a 5 year program of training and research aimed at becoming an independent clinical eating disorders (EDs) researcher, who uses neuroimaging as a tool to test and further develop a model of the parallels between EDs and substance use disorders. To obtain this training, Dr. Broft will build upon previous experience in neuroimaging addictive disorders to generate a collaboration between the Eating Disorders Research Unit, the Division of Substance Abuse and Division of Functional Brain Mapping of Columbia University, the proposed environment for these plans. Dr. Broft's training plan includes (1) development of clinical research expertise in EDs (2) development of expertise in technical aspects of PET neuroimaging (3) training in translational aspects of her research [preclinical reward literature; clinical addictions research] (4) training in responsible conduct of research. Dr. Broft's research plan builds on several lines of evidence which indicate that bulimia nervosa (BN) - an ED characterized by binge eating episodes and loss of control over food - and addictive disorders share overlapping features, are comorbid with one another, and are interrelated. Her proposed work translates PET studies of dopamine (DA) systems, fruitful in addictions research, into the study of BN, and will associate PET outcome measures with behavior. Patients and controls will undergo 2 PET scans with [11C]raclopride, pre/post pharmacological challenge with methylphenidate, for the measurement of striatal DA type 2 (D2) receptor binding potential (BP) and DA release. Patients with BN and controls will then participate in laboratory sessions designed to quantify the reinforcing value of food, using a progressive ratio (PR) computerized work task developed in substance abuse research and adapted for the study of eating disorders. Primary hypotheses include: (i) decreased striatal D2 receptor BP and DA transmission in BN vs. controls, (ii) that patients with BN will 'work' harder for food and consume more food in a PR laboratory task, compared to controls, and (iii) that low striatal DA transmission willbe associated with increased binge size in the behavioral paradigm, similar to addictive disorders. If these abnormalities characterize substance use disorders and BN, this will substantiate their importance, and direct investigations to the factors that contribute to the development of the different pathologies in BN, addictive disorders, and possibly other disorders of motivated behavior. Public health significance: The incidence of BN is 1-2.4% in young adult women, and is highly comorbid with substance use disorders. Treatments for BN are limited, and there is no firm understanding of its pathogenesis. The discoveries on the pathogenesis of addictive disorders offer one of the first viable models for investigation of a neurobiology of BN ..